TensorRT-LLMs/tests/unittest/llmapi/test_llm_models_multi_gpu.py
Kaiyu Xie 2631f21089
Update (#2978)
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
2025-03-23 16:39:35 +08:00

12 lines
398 B
Python

from test_llm_models import llm_test_harness, qwen2_model_path, sampling_params
from utils.util import skip_single_gpu
@skip_single_gpu
def test_llm_qwen2_tp2():
llm_test_harness(qwen2_model_path,
inputs=['A B C'],
references=['D E F G H I J K L M'],
sampling_params=sampling_params,
tensor_parallel_size=2)